michbur / biogram Goto Github PK

degenerate(c("A", "B", "A"), element_groups = list(A = "B", B = "A"))

Should also test if n-grams have proper class.

seq should be first

calc_ed gives too large distances, when the merge inflates number of moves between groups. See

l2 <- list('1' = c("a", "b", "c"),
'2' = c("d", "e", "f"))

l3 <- list('1' = c("a", "b"),
'2' = c("d", "e"),
'3' = "c",
'4' = "f")

calc_ed(l2, l3)

Should be 2, is 3.

decode_ngrams should have also list of vectors output (in addition to the vector of pasted together sequences)

binarize:

ngram_matrix <- as.matrix(ngram_matrix)
ngram_matrix <- ngram_matrix > 0
storage.mode(ngram_matrix) <- "integer"

list2matrix

seqs_m <- tolower(t(sapply(seqs_list, function(i)
c(i, rep(NA, max(lengths(seqs_list)) - length(i))))))